1. Technical Field
The present invention relates to a method for the preparation of probes, in particular preparation of selective probes for the identification of nucleic acids; a matrix comprising such probes, and the use of such a probe for the hybridization of nucleic acids.
2. Background of the Invention
Probes for the hybridization of nucleic acids are used to show the presence of a certain nucleic acid in a test solution. Traditional probes provide no appreciable differences concerning detectable features in free and hybridized conditions but the nucleic acid is determined separation of hybridized and non-hybridized probes (Gillespie and Spiegelman, J. Mol. Biol. 12, 829, 1956). Homogenous test methods, however, use probes the signals of which change at hybridization (FIG. 1). Such probes are composed by a sequence-recognizing part (SID) and a reporting group (RG) (FIG. 2) where SID as a rule is a synthetic oligodeoxy ribonucleic acid (Barton, J., U.S. Pat. No. 5,157,032; Yamana et al., Nucl. and Nucl., 11 (2-4), 383, 1992; Linn et al., EP-A-0 710 668, U.S. Pat. No. 5,597,696) or a nucleic acid analogue (Kubista, PCT/SE97/00953). RG is commonly a colouring agent the fluorescence of which increases at the binding to a nucleic acid. As shown by Nygren et al, Biopol., 46, 39-51 (1998) the properties of such colouring agents are heavily dependent on the sequence of the nucleic acid. This, as shown in PCT/SE97/00953, leads to the fact that both the background fluorescence of free probes as well as the fluorescence of the hybridized probes depend on the sequence of the SID and consequently of the target sequence (MS) selected being complementary to SID. Further, it is probable. that even the sequence closest to MS is of importance.
Nucleic acids can, as a rule, be determined selectively using a great number of probes which differ in the SID part by recognizing different MS which constitute unique segments of the nucleic acid. A nucleic acid having the length m, comprises m+n+1 stretches of the length n which all are potential MS. As the length of MS has not to be particularly long in order to be unique (a 15 bases long stretch appears as an average once per 415=109 bases) a great number of probes can be directed to a certain nucleic acid. A genome having 1000 bases, which is rather characteristic of a virus can be determined using 981 probes having the length of 20 bases, 982 of the length 19, 983 of the length 18, etc. Bacterial genomes which 5 are considerably larger, can be determined using a still greater number of different probes. In order to determine a specific mutation the choice is more restricted as the sequence of the probe has to overlap the presumptive mutation, but still there are several alternatives.
In PCT/SE97/00953 one selects the sequence of SID starting from the knowledge of the known properties of RG, and for asymmetric cyanine dye stuffs SID""s are proposed the terminal bases, preferably, mixed pyrimidines (-TT or -CT). This strategy comprises several restrictions. On one hand detailed studies of the properties of the dye stuff are required which can not be trivially extrapolated to the properties of the probes, due to differences in experimental conditions, etc, and on the other hand the sequence requirements as a rule of an indefinite number of sequences (xe2x85x9 of all sequences, e.g., end with either -TT or -CT). Finally, there is no consideration concerning the sequence closest to MS.
The present invention solves these problems. Basically, it is a method for determining which of a great number of potential probes to a certain nucleic acid, that has the lowest background signal and which of these that obtains the strongest signal at hybridization. The difference in determinations is the increase in signal strength which is obtained using the different probes.